Data Mining for Global Change: Furthering Science, Knowledge
The following is a special contribution to this blog byKarsten Steinhaeuser, a Research Associate in the Department of Computer Science and Engineering at the University of Minnesota involved with a National Science Foundation Expeditions in Computing on Understanding Climate Change: A Data Driven Approach and the Planetary Skin Institute. Karsten describes the Expeditions effort here.
Climate change is a defining environmental challenge facing our planet as rising temperatures, increased severity and frequency of extreme events, and transformation of the global ecosystems are placing unprecedented stress on society, natural resources and man-made infrastructure.
A team of researchers led by Vipin Kumar at the University of Minnesota is exploring ways in which computer scientists can help answer questions surrounding climate change, ecosystem health and global sustainability. The effort is driven by two major initiatives: an NSF Expeditions in Computing on “Understanding Climate Change: A Data Driven Approach” and the GOPHER project, which is affiliated with the Planetary Skin Institute – named as one of Time Magazine’s Best 50 Inventions of 2009 and recently highlighted in The Economist.
The overarching goal of these research activities is to provide innovative, computationally-driven solutions to advance our understanding of the global climate and ecosystems, monitor their current state and improve projections of climate change and its impact on natural and human-made systems. Data driven approaches that have been highly successful in other scientific disciplines hold significant potential for application in environmental sciences. This work addresses key challenges by developing methods that take advantage of the wealth of climate and ecosystem data available from satellite and ground-based sensors, the observational record for atmospheric, oceanic and terrestrial processes, and physics-based climate model simulations. Currently, the focus is on several broad areas including novel methods for analyzing historical climate data, various aspects of modeling tropical cyclone activity, multi-model ensemble methods for evaluating and combining simulation output from multiple climate models, and change detection in space-time data.
In addition to addressing specific science questions, however, these projects also aim to facilitate interaction between computer scientists and researchers in the climate and environmental sciences, foster and strengthen interdisciplinary collaborations and build a community at this interface of computer science and the climate and environmental sciences. For instance, members of the team have been involved in the organization of recent workshops on this topic, including theFirst Workshop on Understanding Climate Change from Data, theFirst International Workshop on Climate Informatics, and the IEEE International workshop on Knowledge Discovery from Climate Data.
Editor’s note: The kinds of successful interdisciplinary collaborations Karsten describes here are consistent with those that NSF is attempting to foster through its variousSEES solicitations described in a separate blog post earlier today.